"After all that, I can only say one of two things: “You’re welcome” or “I’m sorry” for taking you through what the world looks like from my perspective on a day-to-day basis. It’s fun. And frustrating."
I say that your work is very interesting and I look forward to every post. I can't believe that you're able to post not one but two compelling article every week. Where do you get the time?
Question, since I do not have your email address: Can you (or have you) post on the average number of parishioners per minister/priest/whatever per denomination?
That's a great question about ratio of clergy to parishioners. I'm not even sure how I could go about answering it. There may be a handful of denominations that consistently post statistics on their number of ministers. And it would be mostly mainline traditions, which would make the ratios pretty low.
The reason I asked is actually triggered by what's going on in the RC church with their great many parish closing and mergers, combined with their severe clergy shortage. I've read on many diocese and archdiocese sites, e.g. St. Louis Archdiocese, Archdiocese of Seattle and many others, that report priest/parishioner ratios of 1/825-1500! Absolutely astounding numbers.
I suspect that the real-life ratios are probably much smaller as so many registered Catholics are not attending, but even if the numbers were half the problem still exists. This is called clergy burnout.
"that report priest/parishioner ratios of 1/825-1500"
I think Ryan recently posted on this topic. You already noted that we need to factor in attendance. Another aspect worth considering is the number and length of services, it isn't unheard of for a parish to average more than 7 services per week that are each 30 minutes, while the typical Protestant church has 1-2 services that are each ~1 hour long.
I am not sure what responsibilities look like outside of this though. I would expect that preparing the homily is insignificant. My wife does pulpit supply and spends 10 hours to write a typical sermon, but a homily can be much shorter and it wouldn't surprise me if 3 hours of prep was sufficient.
Burnout probably also depends on how many home visits, last rites, weddings, etc they are responsible for per year. I am not sure how typical this is, but my Catholic grandparents never asked for home visits in their last decades of life, and only had last rites done at the very end. From their parish's perspective, they were just names on the member rolls and not much else.
The Catholics would keep some statistics on this. For us Jews, our subdivisions differ a lot. The Conservative and Reform movements have centralized hiring processes for their affiliate congregations. They have guidelines for how many senior and associate rabbis that congregations of different sizes ought to have. There is some autonomy that individual synagogues have, often restrained by finances that affect hiring. And there is currently a shortage of rabbis for those congregations, so that synagogues can now seek clergy trained outside their own movements. The Orthodox world is more amorphous. They have less standardized training, men seeking rabbinical education who have no intention of becomeing congregational rabbis, opting largely for positions in the educational system. This would be much harder to track.
A lot of perception depends on what you experience. The number of auto collisions in a day relative to the number of drivers is pretty small. If you are the claims adjuster receiving multiple calls each day, your mind screens out the people who got where they want to go safely. The number of kids actually chased across a college quad by a hostile student with a bullhorn is pretty small relative to the student body. Even the number of cheering onlookers is not that large. But small numbers, whether fender benders or harassed students, still have public importance. Attention beyond real prevalence still has a public benefit, maybe even a public imperative.
On trend rates, the prevalence of diabetes has accelerated during my professional lifetime as an endocrinologist. The ability to avoid or cope with organ damage from it has greatly improved, again gradually over decades. Neither with statistical inflection points. Both important.
One of the limitations of statistics, whether trend lines that do not change quickly or perceptions that seem exaggerated relative to the raw data, is that they can misrepresent the importance of what is being experienced.
"Whenever I post a graph that points to the fact that a significant majority of Americans will die with the same faith that they were born into (65-70% of us), you know what the first comment will be? 'Well, I grew up a fundamentalist Pentecostal, then converted to Orthodox Judaism, and am now a pantheist who dabbles in Buddhism.'"
As someone who's currently ACNA / raised dispensationalist, I feel like this piece was written especially for me.
Seriously, keep up the good work. To share an anecdote about your influence, one of the prominent laymen in my church once brought up some statistical fact about Christians, and my immediate (and correct) response was "So what you're really telling me is that you read Ryan Burge's Substack".
I honestly can't express how grateful I am for your work. I know enough about statistics to know how maddening this all can be. And enough about the church to understand the denial.
The You Gov example brings up the issue I see over and over again that as a civilisation we are *terrible* at educating people on how prevalent things are in society. This has massive impact on attitudes to things like immigration. Hans Rosling use to speak publicly about this all the time.
If I was a politician in power I would be communicating about these misunderstandings all the time but I imagine they don't because they benefit from many of them.
Also, people aren't very good at mental arithmetic. Looking at the graphs, the proportion Protestants (not shown) must be well below zero, maybe -10 %. That happens a lot
As a data person myself (albeit in the environmental sciences generally, and atmospheric physics in particular) I am very interested in your data mining work. I have a deep appreciation for it. In general the data sets I deal with are much larger than the ones you have to work with, and tend to be unambiguous measurements, or at least with fairly well-understood uncertainties.
One thing that troubles me is that I do believe there is regional variation in the US, on many variables. For those of us trying to make decisions about what to do with and in our churches, given the huge changes in the religious landscape that have happened, we need better information about what is happening, closer to home. The Cooperative Election Study has a wonderful array of variables, but with only 60,000 observations to work with for the whole US, it is tough to get a handle on regional variations. I started trying to do that in part because of the 2024 electoral map: clearly the northeast is not like the rest of the country, looking at the state-level data. I wanted to focus on the "nothing-in-particulars". But I had to aggregate eastern NY, MA, and CT to get a decent number of observations (427 N-Ps out of 1,968) that I thought might be representative. The results were interesting and on the whole, not too surprising. As examples: our N-Ps had higher educational attainment than the US at large, we had a higher proportion of atheists, and a higher proportion of Catholics compared to Protestants.
Three caveats: 1) of course the uncertainty in the regional estimates are larger, although I did experiment with comparing my regional estimates to similarly-sized random samples of the entire data set and 2) we do ministry with real live individuals, not statistics, so we have to be wary of applying what we learn this way to any given person and 3) your (Ryan Burge's) focus has been on temporal trends, and I just worked with one year.
This is a long-winded way of asking if you ever dabble in regional variations ?
I have thought about regional variations, I bit. I think I wrote a post for CT about that subject a long time ago. I remember that evangelicals in New England were not as conservative as evangelicals in the South, for instance.
But it’s a good idea for a new post. I’m putting in my “ideas” note.
Ryan, I share your frustration and concern. In the field of quality management, an assessment of the general public's numerical literacy (see the Red Bead Experiment) generally results in a 100% failure rate.
There are tools that help distinguish Signals (outliers) from Noise. One of the more basic tools is a run chart such as the one you used at the beginning of the article - Support for Abortion (1977-2024). One of the more basic standards for identifying trends is the Rule of 7. Link to an article in response to one of your previous posts: https://timjclark.substack.com/p/a-quality-management-perspective
It is all too common for people to make conclusions based on one or two data points. W. Edwards Deming developed his Red Bead Experiment (see Test Your Knowledge of the variation principle) to help raise awareness of better methods for developing numerical literacy and improving decision-making. Link to an illustration of the experiment and additional information at the following. https://successthroughquality.com/
"After all that, I can only say one of two things: “You’re welcome” or “I’m sorry” for taking you through what the world looks like from my perspective on a day-to-day basis. It’s fun. And frustrating."
I say that your work is very interesting and I look forward to every post. I can't believe that you're able to post not one but two compelling article every week. Where do you get the time?
Question, since I do not have your email address: Can you (or have you) post on the average number of parishioners per minister/priest/whatever per denomination?
Thanks or your work!
That's a great question about ratio of clergy to parishioners. I'm not even sure how I could go about answering it. There may be a handful of denominations that consistently post statistics on their number of ministers. And it would be mostly mainline traditions, which would make the ratios pretty low.
The reason I asked is actually triggered by what's going on in the RC church with their great many parish closing and mergers, combined with their severe clergy shortage. I've read on many diocese and archdiocese sites, e.g. St. Louis Archdiocese, Archdiocese of Seattle and many others, that report priest/parishioner ratios of 1/825-1500! Absolutely astounding numbers.
I suspect that the real-life ratios are probably much smaller as so many registered Catholics are not attending, but even if the numbers were half the problem still exists. This is called clergy burnout.
"that report priest/parishioner ratios of 1/825-1500"
I think Ryan recently posted on this topic. You already noted that we need to factor in attendance. Another aspect worth considering is the number and length of services, it isn't unheard of for a parish to average more than 7 services per week that are each 30 minutes, while the typical Protestant church has 1-2 services that are each ~1 hour long.
I am not sure what responsibilities look like outside of this though. I would expect that preparing the homily is insignificant. My wife does pulpit supply and spends 10 hours to write a typical sermon, but a homily can be much shorter and it wouldn't surprise me if 3 hours of prep was sufficient.
Burnout probably also depends on how many home visits, last rites, weddings, etc they are responsible for per year. I am not sure how typical this is, but my Catholic grandparents never asked for home visits in their last decades of life, and only had last rites done at the very end. From their parish's perspective, they were just names on the member rolls and not much else.
The Catholics would keep some statistics on this. For us Jews, our subdivisions differ a lot. The Conservative and Reform movements have centralized hiring processes for their affiliate congregations. They have guidelines for how many senior and associate rabbis that congregations of different sizes ought to have. There is some autonomy that individual synagogues have, often restrained by finances that affect hiring. And there is currently a shortage of rabbis for those congregations, so that synagogues can now seek clergy trained outside their own movements. The Orthodox world is more amorphous. They have less standardized training, men seeking rabbinical education who have no intention of becomeing congregational rabbis, opting largely for positions in the educational system. This would be much harder to track.
I might watch a show about Cedar Rapids Methodists!
I always glean new ideas that deepen and help me put meaning to my understanding of the information I take in daily. Thank you Ryan!
A lot of perception depends on what you experience. The number of auto collisions in a day relative to the number of drivers is pretty small. If you are the claims adjuster receiving multiple calls each day, your mind screens out the people who got where they want to go safely. The number of kids actually chased across a college quad by a hostile student with a bullhorn is pretty small relative to the student body. Even the number of cheering onlookers is not that large. But small numbers, whether fender benders or harassed students, still have public importance. Attention beyond real prevalence still has a public benefit, maybe even a public imperative.
On trend rates, the prevalence of diabetes has accelerated during my professional lifetime as an endocrinologist. The ability to avoid or cope with organ damage from it has greatly improved, again gradually over decades. Neither with statistical inflection points. Both important.
One of the limitations of statistics, whether trend lines that do not change quickly or perceptions that seem exaggerated relative to the raw data, is that they can misrepresent the importance of what is being experienced.
Seriously, this is an excellent piece. I hope you do more like this about the art (not science) of statistical analysis.
"Whenever I post a graph that points to the fact that a significant majority of Americans will die with the same faith that they were born into (65-70% of us), you know what the first comment will be? 'Well, I grew up a fundamentalist Pentecostal, then converted to Orthodox Judaism, and am now a pantheist who dabbles in Buddhism.'"
lol
As someone who's currently ACNA / raised dispensationalist, I feel like this piece was written especially for me.
Seriously, keep up the good work. To share an anecdote about your influence, one of the prominent laymen in my church once brought up some statistical fact about Christians, and my immediate (and correct) response was "So what you're really telling me is that you read Ryan Burge's Substack".
I honestly can't express how grateful I am for your work. I know enough about statistics to know how maddening this all can be. And enough about the church to understand the denial.
Bless you.
I enjoy the comments on your Substack post versus the craziness of the comments on the Facebook posts.
You and me both, Wesley. But those Facebook posts earn me some real Zuck Bucks at this point. So I can’t stop!
I wish I could post a laughing emoji!
Bravo Ryan! Well done!
The You Gov example brings up the issue I see over and over again that as a civilisation we are *terrible* at educating people on how prevalent things are in society. This has massive impact on attitudes to things like immigration. Hans Rosling use to speak publicly about this all the time.
If I was a politician in power I would be communicating about these misunderstandings all the time but I imagine they don't because they benefit from many of them.
Please rant more. That was interesting.
Oh, you should see the anger at me on X right now for posting a graph that had a faulty weight variable.
My analysis was done exactly the correct way. Code was good. I checked twice.
But the survey team made an error when calculating the weights and it changed the outcome in a major way.
And somehow that is evidence of my deep bias and lack of skill as a data analyst.
Also, people aren't very good at mental arithmetic. Looking at the graphs, the proportion Protestants (not shown) must be well below zero, maybe -10 %. That happens a lot
As a data person myself (albeit in the environmental sciences generally, and atmospheric physics in particular) I am very interested in your data mining work. I have a deep appreciation for it. In general the data sets I deal with are much larger than the ones you have to work with, and tend to be unambiguous measurements, or at least with fairly well-understood uncertainties.
One thing that troubles me is that I do believe there is regional variation in the US, on many variables. For those of us trying to make decisions about what to do with and in our churches, given the huge changes in the religious landscape that have happened, we need better information about what is happening, closer to home. The Cooperative Election Study has a wonderful array of variables, but with only 60,000 observations to work with for the whole US, it is tough to get a handle on regional variations. I started trying to do that in part because of the 2024 electoral map: clearly the northeast is not like the rest of the country, looking at the state-level data. I wanted to focus on the "nothing-in-particulars". But I had to aggregate eastern NY, MA, and CT to get a decent number of observations (427 N-Ps out of 1,968) that I thought might be representative. The results were interesting and on the whole, not too surprising. As examples: our N-Ps had higher educational attainment than the US at large, we had a higher proportion of atheists, and a higher proportion of Catholics compared to Protestants.
Three caveats: 1) of course the uncertainty in the regional estimates are larger, although I did experiment with comparing my regional estimates to similarly-sized random samples of the entire data set and 2) we do ministry with real live individuals, not statistics, so we have to be wary of applying what we learn this way to any given person and 3) your (Ryan Burge's) focus has been on temporal trends, and I just worked with one year.
This is a long-winded way of asking if you ever dabble in regional variations ?
I have thought about regional variations, I bit. I think I wrote a post for CT about that subject a long time ago. I remember that evangelicals in New England were not as conservative as evangelicals in the South, for instance.
But it’s a good idea for a new post. I’m putting in my “ideas” note.
Ryan, I share your frustration and concern. In the field of quality management, an assessment of the general public's numerical literacy (see the Red Bead Experiment) generally results in a 100% failure rate.
There are tools that help distinguish Signals (outliers) from Noise. One of the more basic tools is a run chart such as the one you used at the beginning of the article - Support for Abortion (1977-2024). One of the more basic standards for identifying trends is the Rule of 7. Link to an article in response to one of your previous posts: https://timjclark.substack.com/p/a-quality-management-perspective
It is all too common for people to make conclusions based on one or two data points. W. Edwards Deming developed his Red Bead Experiment (see Test Your Knowledge of the variation principle) to help raise awareness of better methods for developing numerical literacy and improving decision-making. Link to an illustration of the experiment and additional information at the following. https://successthroughquality.com/
You list among the smaller denominations the Presbyterian Church of America. Did you mean Presbyterian Church in America?
I did. It's a mistake I make ALL the time.
What an odd conundrum. You'd think that the Presbyterians would have resolved this by now.